illegal/unrecommended use of IVOA identifiers

[Paul Harrison]

On 06.11.2006, at 08:06, Ray Plante wrote:

> Hi John et al.
>
> I don't know if this has been pointed out to you or not; however,  
> the Plastic Specification is using IVOA identifiers in a non- 
> standard way to identify its messages.
>
> The IVOA Identifiers specification
> (http://www.ivoa.net/Documents/latest/IDs.html) states that the use  
> of the ivo indicates that the identifier has been registered with  
> an IVOA registry (section 3.2.2).
>
> To be compliant, each of the Plastic message identifiers would have  
> to be registered as a separate resource.  While in principle this  
> is possible, this is highly discouraged.  Instead, I would  
> recommend the approach that is to be used by the VOSpace standard  
> in which names are identified using a # suffix; e.g.,
>
>    ivo://votech.org/plastic#info/getIVORN
>
> For use of an identifier of this form to be compliant, only the  
> ivo://votech.org/plastic resource need be registered.  We have a  
> new schema we are putting forward for this purpose called  
> VOStandard (http://www.ivoa.net/internal/IVOA/RegUpgradeSummer2006/ 
> VOStandard-v0.1.xsd).
> It defines two new Resource sub-classes, Standard and StandardService,
> that registers the existance of a standard; the former is probably  
> most appropriate for Plastic.  These types do not currently provide  
> metadata for defining what I might call properties (e.g. info/ 
> getIVORN); however, we should probably add that.  Note that the  
> standard being described need not be an IVOA standard.

I could easily move the types that exist in the VOSpace schema for  
these "properties" into the VOStandard schema, so that they become  
more "generic" - in fact I am about to do some schema work on VOSpace  
so it would be an opportune moment to make the change.

Cheers,	
	Paul.

Paul Harrison
ESO Garching
www.eso.org